MoE expert parallel execution
Parallel execution of 256 routed MoE experts with token-to-expert dispatch and weighted aggregation. Each expert is a SwiGLU MLP (gate_proj, up_proj, down_proj). This operation performs expert masking, token gathering per expert, parallel expert computation, weight application, and scatter-add aggregation back to original token positions.
Reported evidence · last observed 2026-07-10. Reported by source; not independently reproduced. No public source. License unknown.
Current records
Workloadsuite of 16 cases · mean latencyhidden_size = 4096 · n_routed_experts = 256 · num_experts_per_tok = 8 · moe_intermediate_size = 204816 cases
Not measured on H100 for this workload. Challenges →
12.37ms97% SOL1.00×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
22.44ms97% SOL1.03×Reported · license unknown · no source2026-07-15
Measured exactly what you asked. Reported by source; not independently reproduced.
32.56ms96% SOL1.08×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
42.59ms96% SOL1.09×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
52.66ms96% SOL1.12×Reported · license unknown · no source2026-08-21
Measured exactly what you asked. Reported by source; not independently reproduced.
62.85ms95% SOL1.20×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
72.87ms95% SOL1.21×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
82.88ms95% SOL1.21×Reported · license unknown · no source2026-08-15
Measured exactly what you asked. Reported by source; not independently reproduced.
92.93ms95% SOL1.24×Reported · license unknown · no source2026-07-16
Measured exactly what you asked. Reported by source; not independently reproduced.
103.02ms95% SOL1.27×Reported · license unknown · no source2026-08-17
Measured exactly what you asked. Reported by source; not independently reproduced.
113.51ms93% SOL1.48×Reported · license unknown · no source2026-08-19
Measured exactly what you asked. Reported by source; not independently reproduced.
123.62ms92% SOL1.53×Reported · license unknown · no source2026-07-27
Measured exactly what you asked. Reported by source; not independently reproduced.
133.92ms91% SOL1.66×Reported · license unknown · no source2026-07-27
Measured exactly what you asked. Reported by source; not independently reproduced.
143.99ms91% SOL1.68×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.
154.88ms88% SOL2.06×Reported · license unknown · no source2026-07-10
Measured exactly what you asked. Reported by source; not independently reproduced.